Bioorganic & Medicinal Chemistry Letters 17 (2007) 1455–1465

Structure–activity relationship analysis of a novelnecroptosis inhibitor, Necrostatin-5

Ke Wang,a Jinfeng Li,a Alexei Degterev,b Emily Hsu,b Junying Yuanb,�

and Chengye Yuana,*,�

aState Key Laboratory of Bio-organic and Natural Products Chemistry, Shanghai Institute of Organic Chemistry,

Chinese Academy of Sciences, Shanghai 200032, ChinabDepartment of Cell Biology, Harvard Medical School, 240 Longwood Avenue, Boston, MA 02115, USA

Received 15 August 2006; revised 12 October 2006; accepted 7 November 2006

Available online 21 November 2006

Abstract—Necrostatin-5 (Nec-5) is a novel potent small-molecule inhibitor of necroptosis structurally distinct from previouslydescribed Necrostatin-1 (Nec-1), and therefore, represents a new direction for the inhibition of this cellular caspase-independentnecrotic cell death mechanism. Here, we describe a series of structural modifications of Nec-5 and the structure–activity relationship(SAR) of Nec-5 series in inhibiting necroptosis.� 2006 Elsevier Ltd. All rights reserved.

N

OO

OMe

Cell death is traditionally classified either as apoptosisor necrosis. Apoptosis is regulated by an evolutionarilyconserved cellular mechanism that proceeds throughspecific signal transduction pathways common todifferent cell types. Necrosis, on the other hand, isthought to be an unregulated cellular response tooverwhelming stress. Despite the prevalence of necrosisunder pathologic conditions, therapeutic strategies toprevent cell death in pathological conditions havetargeted apoptosis rather than necrosis, because ofthe perception that necrosis is an unregulated andnon-specific process, and therefore, difficult to betargeted for therapeutic purposes.

Apoptosis has been extensively characterized over thepast decade.1 However, there is an increasing awarenessthat apoptosis is not the only regulated cell death mech-anism. For example, although stimulation of the Fas/TNFR death receptor (DR) family triggers a canonical‘extrinsic’ apoptosis pathway, it was demonstrated thatin the absence of intracellular apoptotic signaling, Fas/TNFR is capable of activating a common non-apoptotic

0960-894X/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.

doi:10.1016/j.bmcl.2006.11.056

Keywords: Necroptosis; SAR; Inhibitor; Nec-5; Caspase-independent

cell death; Fused pyrimidinon-thiophene derivatives; Ischemic brain

injury.* Corresponding author. Fax: +86 21 54925379; e-mail:

yuancy@mail.sioc.ac.cn� These two authors contributed equally to this work.

death pathway, which we termed ‘necroptosis.’2–5 Nec-roptosis is a regulated cell death pathway, activatedupon stimulation of FasL/TNFa family of death recep-tor ligands under the conditions when apoptosis isinhibited. Necroptosis is characterized by morphologicalfeatures normally attributed to unregulated necrosis.The existence of a regulated cellular necrotic cell deathmechanism raised the possibility to specifically targetnecrotic component of human diseases. As a first exam-ple, we have used Nec-1 to investigate the pathologicalimportance of necroptosis in ischemic brain injurywhich is known to involve both apoptosis and necro-sis.6,7 We have found that treatment with Nec-1 reducedthe volume of the infarct and ameliorated the neurolog-ical deficits in mouse 2 h middle cerebral artery occlu-sion model. This study points toward the importantcontribution of necroptosis to ischemic tissue injury

NH

NH

S

S N

N

SCH2CN

Nec-1 Nec-5EC50=0.49 M EC50 = 0.24 M

Figure 1. Structure and activity of necrostatins: Nec-1 and Nec-5.

1456 K. Wang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1455–1465

and suggests that small molecule necrostatins may repre-sent a novel class of therapeutically relevant molecules.

To further define necroptosis pathway, we have screeneda chemical library of �100,000 compounds for chemicalinhibitors of the necrotic death of human monocyticU937 cells induced by TNFa and zVAD-fmk,6 whichwas used as an operational definition of necroptosis.This screen resulted in the selection of several necropto-sis inhibitors, including Necrostatin-1 (Nec-1), whichefficiently blocked necroptotic death.6 Here we describeanother novel necrostatin, Nec-5 (Fig. 1). AlthoughNec-5 was selected in a screen in the presence ofzVAD-fmk, similarly to that of Nec-1,6 its action isnot dependent upon pharmacological inhibition of casp-ases. Consistent with the direct activation of necroptosiswhen induction of apoptosis is abolished by geneticinactivation of apoptotic machinery,7–9 Nec-5 preventsthe death of TNFa treated FADD-deficient Jurkat cells,which are unable to activate caspases in response to DRsignaling,10 even in the absence of zVAD-fmk. Becausethe induction of necroptosis in FADD-deficient Jukartcells does not rely on the presence of other chemicals,for example, zVAD-fmk, we used this system to deter-mine that the effective concentration for half-maximumresponse (EC50) for Nec-5 was 0.24 lM, which exceedsthe activity of Nec-1 molecule.

In this communication, we describe the structure–activ-ity analysis of Nec-5 series.

Chemically, Nec-5 is known as 3-p-methoxyphenyl-5,6-tetramethylenothieno[2,3-d]pyrimidin-4-one-2-merca-ptoethylcyanide, however, its method of synthesis has

O

S NH2

OEt

O

S N

N

O

SH

OMe

1

2

a b

dc

Scheme 1. Reagents and conditions: (a) cyanoacetate, S8, Et2NH, EtOH, refl

5–6 h, yield 85%; (c) ethanolic HCl, reflux 12–24 h yield 78%; (d) KOH in 7

N

N

S

O

various substituents

Type 3

Figure 2.

not been reported. Our synthetic protocol is as follows(Scheme 1): on reacting 2-amino-3-carbethoxythiophene(1)11,12 with p-methoxyphenyl isothiocyanate, a thioureaanalog was generated. Cyclization of the latter in ethano-lic HCl provided 2-mercapto-3-p-methoxyphenyl-5,6-tetramethylenothieno[2,3-d]pyrimidin-4-one (2),13 whichgave Nec-5 in 92% yield on reaction with BrCH2CN inthe presence of potassium hydroxide.14

Influence of substituents on Nec-5 activity. In order toinvestigate the structure–activity relationship, our strat-egy for the structure modification of Nec-5 series wasprimarily directed at three parts of the molecule (Fig. 2).

Three types of Nec-5 analogs, representing type 1, type2, and type 3, were generated by changing R, Xn orR1, R2, respectively. Since substitution of CH2CN moi-ety by methyl group on sulfur atom of Nec-5 providedpotent compound, type 2 and type 3 molecules contain-ing both mercaptoethylcyanide and methylthioethermoieties were generated.

Influence of substituent on the sulfur atom of Nec-5: syn-thesis of 2-mercapto-3-p-methoxyphenyl-5,6-tetramethy-len-othieno[2,3-d]pyrimidin-4-ones (3). For the studyof the influence of substituents of sulfur atom of Nec-5on their bioactivities, a series of compounds 3a–x wereprepared by reaction of 2 with RX in the presence ofpotassium hydroxide.

As shown in Table 1, of all the compounds tested, onlyfew types of changes retained activity in the necroptosisassay based on the treatment of FADD-deficient Jurkatcells with TNFa, which is described above.6 It should be

S NH

OEt

O

C

S

NH

OMe

S N

N

O

SCH2CN

OMe

Nec-5

ux 12 h, yield 72%; (b) p-methoxyphenyl isothiocyanate, EtOH, reflux

0% EtOH then BrCH2CN, 1–2 h, yield 92%.

OMe

S CH2CN

various aryl units

various substituents

Type 1

Type 2

Table 1. Structure and activity of compounds 3

S N

N

SR

OOMe

3a-x

Entry Compound R Yielda (%) EC50b (lM) Max protc (%)

1 Nec-5 CH2CN 92 0.24 100

2 3a Me 87 0.24 71

3 3b Et 78 Inactive —

4 3c n-Pr 54 Inactive —

5 3d n-Bu 87 Inactive —

6 3e n-Pent 77 Inactive —

7 3f n-Hex 84 Inactive —

8 3g CH2CH@CH2 88 Inactive —

9 3h CH2C„CH 80 6.08 80.7

10 3i CH2C6H5 92 Inactive —

11 3j CH2(C6H4Me-4) 88 Inactive —

12 3k CH2(C6H4OMe-4) 92 Inactive —

13 3l CH2 (C6H4NO2-4) 85 Inactive —

14 3m CH2COMe 80 Inactive —

15 3n CH2COOMe 79 Inactive —

16 3o CH2CONH2 67 Inactive —

17 3p COMe 91 Inactive —

18 3q COC3H7-n 87 Inactive —

19 3r COC6H5 92 Inactive —

20 3s CH2CH2CN 65 5.28 70

21 3t CH2Cl 45 2.22 85

22 3u CH2NO2 36 Inactive —

23 3v CH2C(O)NH(C6H4CF3-2) 76 Inactive —

24 3w CH2CH(OH)CH3 68 Inactive —

25 3x CH2COOH 65 Inactive —

26 3y Me (sulfoxide) — Inactive —

27 3z Me (sulfone) 87 Inactive —

a Yield% denotes percentage yield in the final reaction of synthesis.b EC50 is the effective concentration for half-maximum response.c Max protection represents max viability obtained in the presence of a compound.

K. Wang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1455–1465 1457

noted that while some of the compounds afforded com-plete 100% protection from necroptosis restoring viabil-ity to control, a number of modifications resulted in notonly change in EC50 values, but also in decrease in thedegree of protection as determined by non-linear regres-sion analysis of the viability data using GraphPad Prizmscientific statistical software package.

Experimental data in Table 1 showed that substitutionof ethylcyanide moiety by methyl group in Nec-5 (3a,entry 2 in Table 1) reserved its activity to a significantextent. On the other hand, further extension of carbonchain on sulfur 3b–3f resulted in the loss of activity.Compound 3s in which CH2CN was replaced byCH2CH2CN retained some activity. Compounds 3hand 3t are still somewhat, albeit less, active, while intro-duction of electron withdrawing group (EWG), forexample, 3i, 3n, and 3u, destroyed the activity of themolecule completely. Overall, these data suggest thatthis position of Nec-5 affords some limited flexibility,that is, ethylcyanide side chain or S-methyl moiety,and the presence of thioether bond is greatly preferred.Oxidation of methylthio group to corresponding sulfox-

ide (3y) or sulfone (3z)15 leads to a complete loss ofactivity.

Influence of N-substituents of pyrimidinone part of Nec-5.For the study of the influence of the aryl substituents,3-aryl-5,6-tetramethylen-othieno[2,3-d]pyrimidin-4-one-2-mercaptoethylcyanide (4) series in which Xn wasintroduced into benzene ring were prepared. Since intro-duction of methylmercapto moiety resulted in substan-tial activity, the synthesis of 2-methylthio-3-aryl-5,6-tetramethylenothieno[2,3-d]pyrimidin-4-one derivatives(5) (Fig. 3) was also pursued.

Synthesis of compounds 4 and 5. To prepare 4, compound1 was reacted with aryl isothiocyanate. Resulting thioureaanalog was cyclized smoothly in ethanolic HCl to form2-mercapto-3-aryl-5,6-tetramethylenothieno[2,3-d]pyrimi-din-4-one.13 The later was reacted with BrCH2CN in thepresence of potassium hydroxide to give 4.

As shown in Table 2, 4a with unsubstituted phenyl ringwas inactive, while introduction of 4-methyl group tothe benzene ring (4g) or replacement of methoxyl with

S N

N

O

S CH2CN

Xn

S N

N

O

S CH3

Xn

4 5

Figure 3.

1458 K. Wang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1455–1465

ethoxy moiety (4d) retained some activity. Further in-crease in Xn size, that is, X = 4-OBn in 4e, eliminatedthe activity, indicating important role of X = 4-OMe inNec-5 binding to the target and showing that increasingthe carbon number of R in OR is not favorable to activ-ity, likely due to steric hindrance. Interestingly, X = 4-OCF3, inactivated the molecule (4t), while X = 4-F,4h, retained significant activity. Meanwhile, the activityof 4j was significantly decreased when 4-F was replacedby 4-Cl and 4-Br derivative (4k) was completely inactive.Compound 4q (X = 3, 4-O2(CH2)) showed good activi-ty, indicating highly restricted nature of the target’sbinding pocket interacting with this part of Nec-5 mol-ecule with high preference toward methoxy group.

Synthesis of compound 5. A series of 2-methylthio-3-aryl-5,6-tetramethylenothieno [2,3-d]pyrimidin-4-one (5)analogs was also prepared by reacting 4 with MeI inthe presence of potassium hydroxide.

As shown in Table 3, while 5b and 5g showed someactivity, albeit significantly lower than corresponding 4analogs (4h and 4p), all of other derivatives were inac-

Table 2. Structure and activity of compounds 4

S N

N

O

4

Entry Compound Xn

1 4a H

2 4b 2-OMe

3 4c 3-OMe

Nec-5 4-OMe

4 4d 4-OEt

5 4e 4-OBn

6 4f 2-Me

7 4g 4-Me

8 4h 4-F

9 4i 3-Cl

10 4j 4-Cl

11 4k 4-Br

12 4l 3,4-Me2

13 4m 3,4-Cl214 4n 3,4-F2

15 4o 2,4-(OMe)2

16 4p 3,4-(OMe)2

17 4q 3,4-O2(CH2)

18 4r 4-SMe

19 4s 2-Me, 4-Cl

20 4t 4-OCF3

tive. These data confirm that ethylcyanide moiety isgreatly preferred over methylthio group in fused pyrim-idone ring.

Influence of substituents on thiophene ring of Nec-5. Forthe study of the influence of substituents on thiophenering of Nec-5, 3-p-methoxyphenyl-5,6-disubstitutedthieno[2,3-d]pyrimidin-4-one-2-mercaptoethylcyanide (6)analogs in which cyclohexyl ring fused to thiophenemolecule was replaced by R1 and R2 were preapred.Synthesis of 2-methylthio-3-p-methoxyphenyl-5,6-disub-stituted thieno [2,3-d]pyrimidin-4-one (7) derivatives wasalso pursued.

Synthesis of compounds 6 and 7. Variation of R1 and R2

led to derivatives with different aliphatic ring. Com-pounds of 6 series were synthesized starting from corre-sponding 2-amino-3-carbethoxythiophenes,11,12 whichwere reacted with p-methoxy-phenylisothiocyanate, toobtain thiourea analog and then cyclized smoothly inethanolic solution saturated with dried hydrogen chlo-ride to form 2-mercapto[2,3-d]pyrimidin-4-one.13 Thelater gave target molecule 6 on reacting with BrCH2CNin the presence of potassium hydroxide (Scheme 2).

As shown in Table 4, 6a, 6b, 6c, and 6d which containhydrogen in the R1 and/or R2 position of 5,6-thiophenering were completely inactive. When R1, R2 are bothmethyl groups (6e), high degree of activity is retained.Limited extension of R1 preserved activity to asignificant extent (6j), while extending R2 position wassignificantly more detrimental: 6f (R2 = Et) displayed

Xn

SCH2CN

Yield (%) EC50 (lM) Max prot (%)

87 Inactive —

58 17.8 51.2

87 1.46 87.7

92 0.24 100

85 0.55 83.8

83 Inactive —

87 Inactive —

80 1.80 48.6

81 0.24 85.1

85 5.10 66.9

90 8.50 55.0

87 Inactive —

91 16.7 45.0

76 5.70 39.0

74 Inactive —

47 Inactive —

52 5.70 57.0

79 0.89 100

55 2.33 70.0

74 Inactive —

81 Inactive —

Table 3. Structure and activity of compounds 5

S N

N

O

SCH3

Xn

5

Entry Compound Xn Yield (%) EC50 (lM) Max prot (%)

1 3a 4-OCH3 87 0.24 71

2 5a 4-OBn 75 Inactive —

3 5b 4-F 64 1.57 61.2

4 5c 4-Br 77 Inactive —

5 5d 3,4-Me2 82 16.7 38.0

6 5e 3,4-Cl2 71 Inactive —

7 5f 2,4-(OMe)2 58 Inactive —

8 5g 3,4-(OMe)2 69 16.7 54.5

9 5h 3,4-O2(CH2) 75 Inactive —

10 5i 3-SMe 67 Inactive —

11 5j 2-Me, 4-Cl 71 Inactive —

12 5k 4-OCF3 84 Inactive —

Table 4. Structure and activity of compounds 6

Entry Compound R1 R2 Yield (%) EC50 (lM) Max prot (%)

1 6a H H 80 Inactive —

2 6b H Me 65 Inactive —

3 6c H Et 70 Inactive —

4 6d Me H 88 Inactive —

5 6e Me Me 91 0.45 100

6 6f Me Et 89 5.26 86.8

7 6g Me n-Pr 72 Inactive —

8 6h Me i-Pr 75 Inactive —

9 6i Me C14H29 71 Inactive —

10 6j Et Me 77 1.08 100

11 6k –(CH2)3– 81 0.45 100

Nec-5 –(CH2)4– 92 0.24 100

12 6l –(CH2)5– 65 0.96 83

13 6m CH2CH2CHCH3CH2 72 Inactive —

14 6n –CH@CHCH@CH– 44 0.18 83.3

15 6o CH2CH2NEtCH2 37 Inactive —

16 6p CH2CH2N(i-Pr)CH2 54 Inactive —

S NH2

OEt

O

S NH

OEt

O

C

S

NH

OMe

S N

N

O

SH

OMe

S N

N

O

SCH2CN

OMe

a b

c

R1

R2

R1

R2

R1

R2

R1

R2

6

Scheme 2. Reagents and conditions: (a) p-methoxyphenyl isothiocyanate, EtOH, reflux; (b) ethanolic HCl, reflux; (c) KOH in 70% EtOH then

BrCH2CN rt, 1–2 h.

K. Wang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1455–1465 1459

EC50 of 5.26 lM and 86.8% protection and compounds6g and 6h were inactive. Thus, experimental data

demonstrate that while R1 and R2 contribute to

compound activity, extension of hydrocarbon chainbeyond methyl at the position 6 and, to a lesserextent, at position 5 is detrimental for activity. Changing

1460 K. Wang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1455–1465

the size of aliphatic ring of compound Nec-5 wasalso investigated. 6k with five-membered ring retainedmost of the activity, while 6l (seven-membered ring) wasless active and compounds 6m and 6o were essentiallyinactive. These data indicated that increased size of thealiphatic ring was inactivating, consistent with theside-chain extension data, from these series. Interestingly,substitution of phenyl ring for the cyclohexane ring (6n)retained most of the compound activity.

Synthesis of compound 7. A series of 2-methylthio-3-p-methoxyphenyl-5,6-disubstituted thieno[2,3-d] pyrimi-din-4-ones were also prepared by the procedure shownin Scheme 2 using MeI instead of BrCH2CN as S-alkyl-ation reagent.

Table 5. Structure and activity of compounds 7

S N

N

O

R1

R2

7

Entry Compound R1 R2

1 7a H H

2 7b Me H

4 7c Me Me

5 7d Et Me

6 7e Me Et

7 7f Me n-Pr

8 7g Me i-Pr

9 7h Me C14H29

10 7i –(CH2)3–

11 7j –(CH2)5–

12 7k CH2CH2CHCH3CH2

13 7l –CH@CHCH@CH–

14 7m CH2CH2NEtCH2

15 7n CH2CH2N(i-Pr)CH2

Table 6. Structure and activity of compounds 8

S N

N

O

8

Entry Compound R

1 8a Et

2 8b n-Pr

3 8c n-But

4 8d n-Pent

5 8e CH2CH@CH2

6 8f CH2C„CH

7 8g CH2C6H5

8 8h CH2(C6H4NO2-4)

9 8i CH2COMe

10 8j CH2NO2

11 8k CH2CH2OH

As shown in Table 5, overall activities of these seriesparalleled those of 6 series, yet were generally lower,consistent with previously generated data for other typesof modifications.

Influence of substituents on sulfur and thiophenering of Nec-5: synthesis of compound 8. Since 6kshowed activity, synthesis of compound 8 analogswas carried out to determine if varying thiophenering will translate into different SAR for the sul-fur moiety. Reacting 2-mercapto-3-p-methoxyphenyl-5,6-trimethylenothieno[2,3-d]pyrimidin-4-one with RXin the presence of potassium hydroxide led to theformation of the corresponding compounds of 8series.

SCH3

OMe

Yield (%) EC50 (lM) Max prot (%)

86 Inactive —

88 Inactive —

92 Inactive —

77 Inactive —

90 Inactive —

92 Inactive —

88 Inactive —

81 Inactive —

92 0.24 97.0

78 Inactive —

74 Inactive —

69 0.24 77

71 Inactive —

66 Inactive —

SR

OMe

Yield (%) EC50 (lM) Max prot (%)

88 Inactive —

87 Inactive —

91 Inactive —

78 Inactive —

76 Inactive —

85 2.49 63.4

84 Inactive —

76 Inactive —

65 Inactive —

71 Inactive —

77 7.14 100

Table 7. Structure and activity of compounds 9

S N

N

SR

OOMe

9

R1

R2

Entry Compound R R1 R2 Yield (%) EC50 (lM) Max prot (%)

1 9a CH2C„CH H H 85 Inactive —

2 9b CH2C„CH Me H 77 Inactive —

3 9c CH2C„CH Me Me 78 4.86 90.3

4 9d CH2C„CH Me Et 79 Inactive —

5 9e CH2C„CH Me n-Pr 81 Inactive —

6 9f CH2C„CH –(CH2)5– 76 Inactive —

7 9g Et Me Me 93 Inactive —

8 9h Et Me Et 90 Inactive —

9 9i Et –(CH2)5– 90 Inactive —

10 9j CH2CH2CN Me Me 49 Inactive —

11 9k CH2CH2OH Me Me 55 7.26 78

12 9l CH2CH2OH Me Et 80 Inactive —

13 9m CH2CH2OH Me n-Pr 64 Inactive —

14 9n Et Me n-Pr 88 Inactive —

15 9o CH2C(O)OH Me Me 51 Inactive —

K. Wang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1455–1465 1461

As shown in Table 6, introduction of the five-memberedring generally maintained previously described SAR.Compounds 8a, 8b, 8c, 8d (R = Et, Pr, Bu, Pent) withthe extension of carbon chain were inactive. Introduc-tion of EWG, that is, R = Bz, CH2COOMe, completelyeliminated activity. It is interesting to note that coordi-nated changes in R groups on thiophene ring resultedin surprising preservation of activity, that is, five-mem-bered ring and methyl group combined together (7i) dis-played higher activity than each change separately 3aand 6k, which may indicate somewhat different topologyof the different Nec-5 analogs in the active centerdepending on the combination of substituents in differ-ent parts of the molecule.

Synthesis of compound 9. Preparation of 2-mercapto-3-p-methoxyphenyl-5,6-disubstituted thieno[2,3-d]pyrimi-

Table 8. Structure and activity of compounds 10 and 11

S N

N

O

10 / 11

R1

R2

Entry Compound R1 R2

1 10a Me Me

2 10b Me Et

3 10c –(CH2)3–

4 10d –(CH2)5–

5 11a Me Me

6 11b Me Et

7 11c –(CH2)3–

8 11d –(CH2)5–

din-4-ones was carried out according to Scheme 2,except RX in the presence of potassium hydroxide wereused as alkylating agents.

As shown in Table 7, compounds 9c and 9k possesssome activity, but are significantly less potent than com-pound Nec-5. Notably, compound 9g was inactive, sug-gesting that reduced size of the thiophene ring did nottranslate into higher degree of flexibility in the toleratedsize of the sulfur substituent.

Influence of substituents on thiophene ring and N-pyrimid-inone part. Influence of the substituents of Nec-5 wasstudied by changing thiophene ring and pyrimidinonepart together. Since compounds 4h, 4q, 6n, and 6eshowed substantial activity, synthesis of their derivativeswas pursued.

SCH2CN / CH3

F

Yield (%) EC50 (lM) Max prot (%)

86 4.43 87.6

84 Inactive —

89 0.89 88.7

88 Inactive —

91 2.60 69.2

90 Inactive —

81 3.00 95

74 Inactive —

Table 9. Structure and activity of compounds 12, 13, and 14

S N

N

O

SCH2CN / CH3

O

O

S N

N

O

SCH2CN

O

O

12 / 13

R1

R2

14

R1

R2

Entry Compound R1 R2 Yield (%) EC50 (lM) Max prot (%)

1 12a Me Me 91 1.65 100

2 12b Et Me 89 1.90 100

3 12c –(CH2)3– 86 1.18 100

4 13a Me Me 92 1.11 67.0

5 13b Et Me 93 Inactive —

6 13c –(CH2)3– 90 Inactive —

7 14a –(CH2)3– 91 0.25 100

8 14b –(CH2)4– 90 0.22 100

9 14c –CH@CH–CH@CH– 85 0.15 100

10 14d Me Me 93 0.25 100

1462 K. Wang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1455–1465

Synthesis of compounds 10 and 11. The 3-p-fluorophenyl-5,6-disubstituted thieno[2,3-d]pyrimidin-4-one-2-merca-ptoethylcyanide 10 and corresponding methylthioether11 were generated through reacting mercapto derivativeswith BrCH2CN or MeI, respectively, in the presence ofpotassium hydroxide.

As shown in Table 8, the introduction of fluorine atomat the phenyl ring, where seven-membered ring-contain-ing molecules completely lacked activity (10d and 11d),in contrast to methoxy analog (6l). The latter result isreminiscent of the lack of activity displayed by com-pound 7j. These results point that all three major moie-ties, targeted by our analysis (Fig. 2), make importantcontributions to binding, and multiple unfavorablechanges result in synergistic loss of activity indicativeof the inability of the resulting molecules to properlyoccupy the binding pocket.

Synthesis of compounds 12, 13, and 14. Since 3-(3 0,4 0)-methylene-dioxyphenyl-5,6-tetramethylenothieno [2,3-d]pyrimidin-4-one-2-mercaptoethylcyanide (4q), in whichdioxalane ring is attached to the phenyl moiety, showedsignificant activity, synthesis of its derivatives (13 and 14)

Table 10. Structure and activity of compounds 15 and 16

S N

N

O

15 / 16

R1

R2

Entry Compound R1 R2

1 15a Me Me

2 15b CH2CH2CH(CH3)CH2

3 15c –(CH2)3–

4 16a Me Me

5 16b CH2CH2CH(CH3)CH2

6 16c –(CH2)3–

was carried out. Synthesis of 3-(3 0,4 0)-methylene-dioxy-phenyl-5,6-disubstituted thieno[2,3-d]pyrimidin-4-one-2-mercaptoethylcyanide (12) and corresponding methyl-thio ether (13) was performed through S-alkylation withBrCH2CN or MeI in the presence of potassium hydrox-ide, respectively. And synthesis of 3-(3 0,4 0)-ethylene-dioxyphenyl-5,6-disubstituted thieno[2,3-d]pyrimidin-4-one-2 0-mereaptoethylcyanide (14) was performed bythe usual procedure of S-alkylation of correspondingmercapto compounds.

Experimental data in Table 9 showed that withexception of inactive 13b and 13c, all of the compoundsinvestigated either with methylene or with ethylenedioxy moiety attached to phenyl ring on pyrimidonenitrogen atom inhibited enhanced activity, particularlyfor compounds of 14 series. In the latter case, consistentwith our previous conclusion the Xn moiety affordedsignificant flexibility to the structure of the thiophenering.

Synthesis of compounds 15 and 16. Synthesis of 15 and 16analogs was performed in order to study the influence ofsubstituents of benzene ring on activity (Table 10).

SCH2CN / CH3

Yield (%) EC50 (lM) Max prot (%)

87 2.79 90.7

81 16.7 38.0

87 Inactive —

78 Inactive —

85 Inactive —

88 Inactive —

Table 11. Structure and activity of compounds 17 and 18

S N

N

O

SCH2CN / CH3

Xn

17 / 18

Entry Compound Xn Yield (%) EC50 (lM) Max prot (%)

1 17a H 79 NDa NDa

2 17b 4-F 77 3.06 —

3 17c 4-OEt 82 Inactive —

4 17d 3,4-O2(CH2) 82 0.27 100

5 17e 4-OCF3 83 Inactive —

6 17f 4-NMe2 67 Inactive —

7 18a H 82 NDa NDa

8 18b 4-F 78 3.06 —

9 18c 4-OEt 81 Inactive —

10 18d 3,4-O2(CH2) 88 Inactive —

11 18e 4-OCF3 76 Inactive —

a Not determined.

K. Wang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1455–1465 1463

Interestingly, compound 15a showed higher activitythan 4l, proving a first example of coordinated changesin the left and right part of the molecules displayingcompensatory, rather than synergistic effect. It is possi-ble that 3,4-Me-substituted molecule may assume alter-native binding position in the presence of the smaller R1/R2 substituents, resulting in retention of the activity.However, this effect is limited to a particular combina-tion of R1/R2 as 15b and 15c are essentially inactive (Ta-ble 11).

Synthesis of compounds 17 and 18. Since compound 6nshowed good activity, synthesis of its analogs with var-ious phenyl ring substituents (17 and 18) was carriedout, by reacting 2-amino-benzo[b]thiophene-3-carboxyl-ic acid ethyl ester and arylisothiocyanate with NaOH inDMF to generate corresponding mercapto compoundand, subsequently, S-alkylation with BrCH2CN orMeI, respectively, in the presence of potassium hydrox-ide to generate target molecules.

Table 12. Structure and activity of compounds 19 and 20

S N

N

O

S19 / 20

Entry Compound Xn

1 19a 4-OEt

2 19b 4-OBn

3 19c 4-OCF3

4 19d 4-NMe2

5 20a 4-OEt

6 20b 4-OBn

7 20c 4-OCF3

8 20d 4-NMe2

Compound 17d is a potent inhibitor along with 6n and7l consistent with previously defined SAR for othertypes of thiophene ring substituents. Therefore, substitu-tion of phenyl for the cyclohexane ring does not appearto significantly change Nec-5 activity.

Synthesis of compounds 19 and 20. Since methyl groupsin R1 and R2 positions showed significant activity (com-pound 6e), analogs with additional phenyl ring substitu-tions (Xn, compounds 19 and 20) were prepared byreacting corresponding mercapto derivative withBrCH2CN or MeI, respectively, in the presence of potas-sium hydroxide.

Analysis of analogs of 19 and 20 as well as previouslydescribed derivatives with R1 = Me, R2 = Me (Table12) suggests that such modifications are not favorablefor activity, with all the analogs studied being generallyless active than the corresponding cyclohexane moiety(R1, R2 = –(CH2)4–) analogs of Nec-5. Furthermore,

CH2CN / CH3

Xn

Yield (%) EC50 (lM) Max prot (%)

87 7.77 97

79 Inactive —

81 3.70 63

83 Inactive —

90 Inactive —

84 Inactive —

87 Inactive —

84 Inactive —

Table 13. Structure and activity of compounds 21

S N

N

O

SCH2CN

XnR1

R2

21

Entry Compound R1 R2 Xn Yield (%) EC50 (lM) Max prot (%)

1 21a H H H 85 Inactive —

2 21b –(CH2)3– H 90 Inactive —

3 21c –(CH2)3– 4-OEt 67 Inactive —

4 21d –(CH2)5– H 87 Inactive —

5 21e CH2CH2CH(CH3)CH2 3,4-(OMe)2 78 Inactive —

6 21f –(CH2)3– 4-OCF3 88 3.70 70

7 21g –(CH2)5– 4-OCF3 85 Inactive —

8 21h –(CH2)3– 3,4-(OMe)2 71 Inactive —

1464 K. Wang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1455–1465

unlike results obtained with 3-substituents (series 14/15),again, synergistic loss of activity was observed for unfa-vorable changes in thiophene ring and in 4-position, forexample, compounds 4d and 6e showed significantlyhigher activity compared to 19a.

Synthesis of compound 21. For the study of the influenceof the combination of the substituents on thiophene ringand N-pyrimidinone of Nec-5, derivatives of 3-aryl-5,6-disubstituted thieno[2,3-d]pyrimidin-4-one-2- merca-ptoethylcyanide (21) were prepared.

Table 14. Structure and activity of compounds 22

S N

N

O

SR

Xn

22

Entry Compound R Xn Yield (%)

1 22a CH2CH@CH2 H 58

2 22b CH2Ph H 87

3 22c CH2C6H4NO2 H 83

Table 15. Structure and activity of compounds 23

S N

N

O

R1

R2

23

Entry Compound R R1

1 23a Me H

2 23b Me CH2

3 23c Et

4 23d CH2C„CH Me

5 23e CH2C„CH Et

6 23f Me

7 23g Me

8 23h CH2CH2OH Et

9 23i Me C

10 23j Me

As shown in Table 13, combining changes to thiopheneand phenyl rings was detrimental to activity, as onlycompound 21f showed some, yet greatly reduced,activity.

Influence of substituents of sulfur and N-pyrimidinone ofNec-5. For the study of the influence of substituentson sulfur and N-pyrimidinone of Nec-5, 2-mercapto-3-aryl-5,6-tetramethylenothieno[2,3-d]pyrimidin-4-ones (22)which combined unsubstituted phenyl ring and varioussubstituents on sulfur were prepared.

As shown in Table 14, all the derivatives are inactive, con-sistent with the requirement for Xn = 4 0-OMe and prefer-ence for 2-mercaptoethylcyanide moiety as R group.

Influence of substituents of sulfur, N-pyrimidinone as well asthiophene ring. For the study of the influence of simulta-neous substitution on sulfur in thiophene ring and N-pyri-midinone of Nec-5, 2-mercapto-3-aryl-5,6-disubstitutedthieno[2,3-d] pyrimidin-4-ones (23) were synthesized.

As shown in Table 15, consistent with the preference forR1, R2 = –(CH2)4–, R = CH2CN and Xn = 4 0-OMe, allthe analogs of 23 were completely inactive.

SR

Xn

R2 Xn Yield (%)

H H 90

CH2CH(CH3)CH2 3,4-(OMe)2 88

–(CH2)5– H 89

Me 3,4-O2(CH2) 69

Me 3,4-O2(CH2) 79

–(CH2)3– 4-OEt 84

–(CH2)3– 3,4-O2(CH2) 85

Me 3,4-O2(CH2) 79

H2CH2NEtCH2 H 88

–(CH2)3– 4-OCF3 81

K. Wang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1455–1465 1465

Our preliminary SAR study demonstrated that the EC50

value for inhibition of necroptosis in FADD-deficientJurkat T cells treated with TNFa of Nec-5 is closelyrelated to the chemical structure of the molecule. Firstof all, the presence of thioethylcyanide moiety on thea-position of fused pyrimidone-4 part is essential, substi-tution of this moiety results in complete loss of activity.The exceptions are corresponding methylthio ethers as3a and 7i exhibit same EC50 value as Nec-5, althoughit provides significantly lower, max protection value of71% and 97%. Meanwhile, oxidation of the sulfur atom,either to sulfoxide 3y or to sulfone 3z completely elimi-nates the activity. Second, presence of –OMe group inthe para-position of the benzene ring located on pyrim-idone nitrogen is also important. Compound with para-bromophenyl group exhibited (4k) loss of the activitysince variation of the electronic effect of the aryl substit-uents including modification of –OMe group positiongave only less even inactive compound. Compound withpara-fluorophenyl group (4h) displayed significant EC50

value and a slightly decreased max protection of 85.1%,while larger halides were not tolerated. Furthermore,ethylene dioxy group is preferable to methoxy with 14cshowing almost twofold increase in activity. Finally,cyclopentyl (6k), cycloheptyl (6l), and even benzene ring(6n) exhibit certain degree of activity. It is worthy topoint out that introduction of two methyl groups tothe a and b position of thiophene ring will bring com-pound with significant activities. Our results suggest thatwhile Nec-5 displays a stringent SAR, there are alsopositions in the molecule especially phenyl ring attachedto the N-pyrimidone, which can be potentially furtheroptimized to generate more active Nec-5 analogs.

Acknowledgments

This Project was supported in part by the Chinese Acad-emy of Sciences, National Institute of General MedicalSciences (USA), and National Institute of NeurologicalDisorders and Stroke (USA) (to J.Y.) and the NationalNatural Science Foundation of China (Nos. 20272075,20372076 to C.Y.)

References and notes

1. (a) Degterev, A.; Boyce, M.; Yuan, J. Oncogene 2003, 22,8543; (b) Yuan, J.; Lipinski, M.; Degterev, A. Neuron2003, 40, 401.

2. Vercammen, D.; Brouckaert, G.; Denecker, G.; Van deCraen, M.; Declercq, W.; Fiers, W.; Vandenabeele, P.J. Exp. Med. 1998, 188, 919.

3. Matsumura, H.; Shimizu, Y.; Ohsawa, Y.; Kawahara, A.;Uchiyama, Y.; Nagata, S. J. Cell Biol. 2000, 151, 1247.

4. Holler, N.; Zaru, R.; Micheau, O.; Thome, M.; Attinger,A.; Valitutti, S.; Bodmer, J.; Schneider, P.; Seed, B.;Tschopp, J. Nat. Immunol. 2000, 1, 489.

5. Kawahara, A.; Ohsawa, Y.; Matsumura, H.; Uchiyama,Y.; Nagata, S. J. Cell Biol. 1998, 143, 1353.

6. (a) Degterev, A.; Huang, Z.; Boyce, M.; Li, Y.; Jagtap, P.;Mizushima, N.; Cuny, G. D.; Moskowitz, M.; Yuan, J.Nat. Chem. Biol. 2005, 2, 112; (b) Teng, X.; Degterev, A.;Jagtap, P.; Xing, X.; Choi, S.; Denu, R.; Yuan, J.; Cuny,G. D. Bioorg. Med. Chem. Lett. 2005, 15, 5039.

7. (a) Lo, E. H.; Dalkara, T.; Moskowitz, M. A. Nat. Rev.Neurosci. 2003, 224, 29; (b) Gwag, B. J.; Lobner, D.; Koh,J. Y.; Wie, M. B.; Chio, D. W. Neuroscience 1995, 68, 615;(c) Rosenbaum, D. M.; Gupta, G.; D’Amore, J.; Singh,M.; Weidenheim, K.; Zhang, H.; Kessler, J. A.J. Neurosci. Res. 2000, 61, 686; (d) Martin-Villalba, A.;Herr, I.; Jeremias, J.; Hahne, M.; Brandt, R.; Vogel, J.;Schenkel, J.; Herdegen, T.; Debatin, K. M. J. Neurosci.1999, 19, 3809; (e) Martin-Villalba, A.; Hahne, M.;Kleber, S.; Vogel, J.; Falk, W.; Schenkel, J.; Krammer,P. H. Cell Death Differ. 2001, 8, 679.

8. Li, M.; Beg, A. A. J. Virol. 2000, 74, 7470.9. (a) Lin, Y.; Choksi, S.; Shen, H. M.; Yang, Q. F.; Hur, G.

M.; Kim, Y. S.; Tran, J. H.; Nedospasov, S. A.; Liu, Z. G.J. Biol. Chem. 2004, 279, 10822; (b) Wilson, C. A.;Browning, J. L. Cell Death Differ. 2002, 9, 1321.

10. Chan, F. K. J. Biol. Chem. 2003, 278, 51613.11. (a) Gewald, K.; Schinke, E.; Bottcher, H. Chem. Ber 1966,

99, 94; (b) Tranberg, C.; Zickgraf, A.; Giunta, B. N.;Luetjens, H.; Figler, H.; Murphree, L. J.; Falke, R.;Fleischer, H.; Linden, J.; Scammells, P. J.; Olsson, R. A.J. Med. Chem. 2002, 45, 382; (c) Gutschow, M.; Kuerschner,L.; Neumann, U.; Pietsch, M.; Loser, R.; Koglin, N.;Eger, K. J. Med. Chem. 1999, 42, 5437; (d) Gewald, K.;Neumann, G. Chem. Ber. 1968, 101, 1933.

12. (a) Sabnis, R. W. Sulfur Rep. 1994, 16, 1; (a) Sabnis, R.W.; Rangnekar, D. W. J. Heterocycl. Chem. 1999, 36, 333.

13. (a) Vishnu, J. R.; Hrishi, K. P.; Arnold, J. V.J. Heterocycl. Chem. 1981, 18, 1277; (b) Devani, M. B.;Shishoo, C. J.; Pathak, U. S.; Parikh, S. H.; Saha, G. F.;Padhya, A. C. J. Pharm. Sci. 1976, 65, 660; (c) Leistner, S.;Gutschow, M.; Wagner, G. Synthesis 1987, 466; (d)Modica, M.; Santagati, M.; Santagati, A.; Russo, F.;Cagnotto, A.; Goegan, M.; Mennini, T. Bioorg. Med.Chem. Lett. 2000, 10, 1089; (e) Duval, E.; Case, A.; Stein,R. L.; Cuny, G. D. Bioorg. Med. Chem. Lett. 2005, 15,1885.

14. Nec-5: mp 212–214 �C. 1H NMR(CDCl3) (d): 1.82–1.90(m, 4H, CH2), 2.77–2.97 (m, 4H, CH2), 3.89 (s, 5H,OCH3 + CH2CN), 7.18 (d, J = 8.7 Hz, 2H, OCH3–PhH),7.36 (d, J = 8.7 Hz, 2H, OCH3–PhH). IR (KBr, cm�1):2939, 2848, 2249(C„N), 1630(s, C@O), 1444, 1321, 1262,872. MS m/z (rel intensity): 383 (M+) (31.21), 311 (27.60),146(base). Anal. Calcd for C19H17N3O2S2: C, 59.51; H,4.47; N, 10.96. Found: C, 59.13; H, 4.48; N, 10.80.

15. Oxidation of compound 3 was carried out as follows: amixture of 3a (1 mmol) and m-chloroperoxybenzoic acid(m-CPBA) (2.2 mmol) in CH2Cl2 (20 ml) was stirred for48 h. The reaction was completed as monitored by TLC.The product was separated by silica gel column chroma-tography using chloroform as eluent. The product 3z wascrystallized from chloroform–ethanol as colorless crystal,yield 87%, mp 230 �C. 1H NMR(d): 1.86–1.92 (m, 4H,CH2), 2.47 (s, S(O)2CH3), 2.73–2.78 (m, 2H, CH2), 2.91–2.97 (m, 2H, CH2), 3.87 (s, 3H, OCH3), 7.02 (d,J = 9.0 Hz, 2H, ArH), 7.19 (d, J = 9.0 Hz, 2H, ArH).IR(KBr)cmax: 3199, 2938, 2837, 1712, 1655 (s, C@O),1510, 1246, 829. MS m/z (rel intensity): 390 (M+), 359, 311(base), 199, 159. Anal. Calcd for C18H18N2O4S2: C, 55.37;H, 4.65; N, 7.17. Found: C, 55.98; H, 4.79; N, 6.84.